The trick is to model the resonant cavity with the coatings capability. See Documents\Zemax\Samples\Sequential\Polarization\Fabry-Perot.ZMX for an example.

- Mark 

Hello Mark, good to see you here :)

I am trying to model it in NonSeq mode, because it is part of an instrument and I would need to simulate the signal in the detector. 

So far I have put two rectangular volumes to create the cavity and played around with the coatings for Front/Rear faces but the result is that all wavelengths get through. So the physics of resonance is not happening. Im doing something wrong

Any tips ?

thx in advance

Hello Javier

It should work with coatings in non-sequential mode. When you run your raytrace, you have to tick 'Use Polarization'. I have attached an example. I converted the file '\Zemax\Samples\Sequential\Polarization\Fabry-Perot.ZMX' to non-sequential.

If it doesn't work, feel free to open a case with us and we can have a look at your file.

Just as an extra comment, a Fabry Perot can be modelled with coatings but it can't be modelled using Coherent Irradiance.

Sandrine

Dear Sandrine, 

thanks a lot for your input. Getting those rings in NonSeq is something !!

I have been checking your file and I have the following comments:

- It is weird that the results do no react when I modify the coatings from 'FP' to any other.

- The same happens when I change the wavelength or other parameters of the system.

So in general it doesnot react to any change. As you know, a Fabry-Perot filters wavelength by fine-tuning the distance between mirrors, and in the zemax sample files the distance between the two plates is zero. that is the point I dont see. Ideally , I would like to simulate in OS something like this web simulator:

https://lightmachinery.com/optical-design-center/etalon-designer/

Should I open a case or shall we follow here?

Thanks,

Javier

Hi,

Look at the FP coating that is used in the Fabry-Perot sample file:

Coating Name: FP, 3 layer(s)

              Material    Thickness     Absolute    Loop                Taper

                  ALUM       0.004000           0           0                       

                   AIR     150.000000           0           0                       

                  ALUM       0.004000           0           0                       

So this coating is made of a thin .004 wavelength layer of aluminum, a 150 wave thickness of air, and another thin layer of aluminum. The FP is therefore comprised of two pieces of glass in the editor, with Al/Air/Al sandwiched between. In order to tune the etalon, change the thickness of the air.

The only limitation to the FP modeling is that the thickness of the layers does not draw on the layout plots. That's in common with the whole coating capability. Other than that, it will react to wavelength and angle changes just fine: as long as 'Use Polarization' is turned on so that coating effects are considered.

Sandrine, I think the issue with Coherent Irradiance isn't the FP: it's the paraxial lens model in the NS mode. It doesn't consider phase at all. Replace the paraxial lens with a real imaging system and it will work fine in Coherent as well as Incoherent modes. 

- Mark

Hello Mark,

thanks for the reply. I have been looking into it and I have the following comments:

- indeed it reacts to changes in the primary wavelength. Still havent changed the thickness of Air

- the Transmission vs Wavelength is flat to 1 in the NonSeq model, I dont know the reason

- In your FP 'sandwhich', ALUM is defined before as a material (MATE) with = 0.55  0.7  -7.0  . So my next question is, where is the reflectivity of this aluminum defined?. The reflectivy defines then the Finesse and in general the performance of the FabryPerot experiment

Thanks a lot 

Hi Javier

- the Transmission vs Wavelength is flat to 1 in the NonSeq model, I dont know the reason

Make sure 'Use Polarization' is on. If that doesn't fix it, please post your file as a ZAR so we can see what's happening.

- In your FP 'sandwhich', ALUM is defined before as a material (MATE) with = 0.55  0.7  -7.0  . So my next question is, where is the reflectivity of this aluminum defined?. The reflectivy defines then the Finesse and in general the performance of the FabryPerot experiment

The reflectivity is computed from the complex index of the coating layers, the number of coatings, their thickness etc. The reflectivity is a function of the whole stack. If you want to see the reflectivity of the Al layer, put it on a flat piece of glass and use Analysis...Reflectivity.

- M

Hi Javier

I have updated the non-sequential file:

• In the Transmission vs Wavelength, I have modified the settings to plot the front face of the 3rd object (check the settings). 



• I have changed the paraxial lens by a real lens.

Sandrine

And thanks a lot Mark for your help!

Hello Sandrine and Mark,

I start to believe that my issues are more related to the Detector Viewers than the FP itself,, which is already tricky. Here I drop some comments:

- The default sandwich of coatings in None-FP-FP-None. I have noticed that if I remove one FP it behaves different:

None-FP-None-None (ie, modify surface 3, Front) gives a (more or less gaussian) spot of light, with no interference at all. Whereas...

None-None-FP-None (ie, modify surface 2, Back) gives rings so interference is happening.

Does it mean that I dont need to apply the coating on both , and only the Front side of the second mirror is enough ?

- There must be something that stays in memory so that Detector Viewers show incorrect signal. If I try to recover the default rings at 0,55 um , it shows something different. The same happens when I change to different wavelengths.

Even if I reopen the file, it opens several Viewers that I did close (and did not save) before !. The only way is to shut down OS and launch it again.

Sandrine, I still haven't looked at your file with a real lens. I will give you feedback asap

Thanks a lot,

JMV 

Hi Javier

You only need one FP coating on the 2nd window. This is because in non-sequential, we have a rule called the nesting rule that says :


'If a ray strikes more than one object at the exact same point in space; the last object listed in the NSC Editor determines the properties of the surface or volume at that point.'

So actually in non-sequential, the software will only see the FP coating from the second window.

Check in the help file here:

Let me know if that helps.

Sandrine

Thanks for thereply Sandrine. Clearly the 'nesting rule' applies there.

Im still fighting to get my etalon , in my caso made of HfO2 and SiO2 for the UV. Seems to be quite standard mix for the UV. I will keep you posted

thanks,

JMV

Hi,

Very useful discusion.

I managed to create the FP behavior of a coverslip just by using an AIR rectangular volume and use the specs of the slip (in this case 250 micron of silica) as coating of the that AIR rectangular volume. Note that the rectangular volume does not affect the ray tracing, it is just a holder. 

I hope this helps 

Best, Victor

Hello Mark and Sandrine,

I am in the process of modeling a FP etalon system and I have used both sequential and non-sequential modes implementing the etalon as a coating based on the zemax example file and as described above  The results for this idealized case seem reasonable.

A useful simulation tool for my problem would be an opto-mechanical model where I can see the effect of tilt and stress on the etalon plates.  I have tried to follow the approach above from Javier using an 80% reflective coating on each of the air spaced etalon plates but this does not seem to be working.  Perhaps it is just not possible to trace enough rays.  If this approach did work, I could tilt one of the plates , add curvature or a TIRR or Zernike sag surface to represent deformation to study these effects and develop an error budget for my system.

In the past, I have used Solidworks simulation and Sigfit to combine opto-mechanical effects from FEA results into my Zemax model.   Are there any tricks you can recommend to develop an etalon model that would lend itself to this type of analysis?  Can tilt or irregularity be introduced in the coating to simulate these effects?

Thanks,

Pat

I think using tapered coatings would be the best (only?) way to do this. Checkout the TAPR data section in the coating definition file.

Thanks Mark.  I will investigate tapered coatings.  I have no experience in this area, so hopefully I can figure it out from your examples.

Regards,

Pat